This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Abstract Flow cytometry is now a standard analysis platform for diagnostic analysis of human blood samples, primarily through the use of immunophenotyping via cell surface marker labeling with fluorescently-tagged antibodies. Relevance of these applications covers a wide range of areas including leukemia/lymphoma diagnosis and patient specific management, infectious disease, transplant medicine (stem cell enumeration/ characterization), paroxysmal nocturnal hemoglobinuria (PNH) diagnosis and fetal hemoglobin detection. Although such analysis is routine, there are still problems that are amenable to improve flow instrumentation. In this collaboration with Dr. Charles Goolsby's clinical immunophenotyping laboratory, we will use the unique instrumentation developed in Projects 1 and 3 to directly address two of these problems: separation of overlapping fluorophores in multi-color flow analysis which are now routinely six-eight color and loss of subsets of white cells during red cell lysis procedures. The improved spectral resolution instrument developed in Project 3 will be used to determine if deconvolution of complete emission spectra from multiply-stained cells can improve the resolution and quantitation of different cell types as compared to the standard use of optical filters and a complex compensation matrix. The in-line sample preparation device developed in Project 1 will be used to determine if acoustic field separation of red and white cells in a flowing sample stream can eliminate the need for a red cell lysis step with its resultant loss of certain subsets of white cells, particularly fragile abnormal cells. The availability of several types of clinical samples through this collaboration will directly test the utility of these two instruments to address limitations of conventional flow cytometry in a real-world situation. Background Dr. Charles Goolsby is the Floyd Elroy Patterson Professor of Pathology and Director of the Flow Cytometry Clinical and Core Facilities at Northwestern University, specializing in the investigation of the basic cell biology of B chronic lymphocytic leukemia (CLL) using flow cytometry (1-3). Dr. Goolsby has lead also the development of complex, multiparametric flow cytometry based clinical analyses both for diagnostic purposes and for patient specific therapeutic decision/monitoring purposes including assessment of signal transduction pathways in the setting of kinase inhibitor therapies (4). Routine utilization of six to eight color analyses are employed in the clinical laboratory setting. Although flow cytometry is the 'gold standard'for the immunophenotypic analysis of human blood samples for a wide variety of diseases, there are at least two areas in which technical advances would improve the utility of this technology for diagnostics and therapy monitoring. The first concerns the use of multiple, overlapping fluorochromes linked to antibodies to label numerous cell types in a single sample. A complex compensation algorithm on the flow cytometer software estimates the relative contribution of the overlapping fluorochromes, and there is no way to directly test the success of the compensation matrix for individual samples. The upgraded full spectral resolution cytometer to be developed in Project 3 will be used to address this limitation of conventional flow cytometers. A second problem is the requirement to lyse red cells in blood cell samples prior to flow analysis. Although red blood cell lysis protocols are far less selective than gradient separation techniques, in some patients selective loss of specific cell subsets can still be seen, a complication that it is difficult to track for individual samples in a clinical setting. The selective loss of subsets of white cells is a particular problem in the analysis of some acute leukemias, large cell lymphomas, and in specimens with a high apoptotic index where the abnormal, cancer cells can be especially fragile and lost during the red cell lysis procedure. This inconsistent, and difficult to predict on a patient specific sample basis complicates the use of flow cytometry for diagnosis and therapy monitoring. The acoustic instrument developed in Project 1 will be used to determine if specific cell subsets in peripheral blood samples can be separated by field-based manipulation prior to analysis, negating the need for a red blood cell lysis step. Approach Improved Spectral Separation of Multiply Stained Blood Cells The upgraded full spectral resolution cytometer will be used in this collaboration to determine if the separation of overlapping fluorochromes can be improved by measuring the complete emission spectrum of the cell. Preliminary work with fixed clinical samples from Dr. Goolsby's laboratory has shown that the current spectral instrument can collect complete emission spectra of individual cells at the two excitation wavelengths commonly used for multi-color clinical analysis. We have also demonstrated that we can generate standard list-mode data files from large numbers of complete emission spectra collected on a single cell basis. We will use both Fourier transform and wavelet algorithms for deconvoluting the emission spectra into its component emission spectra from the individual fluorochromes, using cells stained with each of the individual cell surface markers, along with unstained cells, as a basis set of control spectra. The intensity of signal from each component of the deconvolved spectra will be compared to those obtained from standard compensation matrix analysis of the same clinical samples as analyzed in the Goolsby laboratory. As a further control, we will set 'virtual filters'on the emission spectra acquired at Los Alamos, using the same wavelength ranges as employed using actual bandpass and longpass filters on the standard flow cytometer used to acquire the clinical data. These data will then be run through the same compensation matrix as used on the standard data in order to determine if the spectral instrument faithfully replicates the emission data measured on the standard cytometer. We will also measure the samples on our in-house commercial instrument using the same settings as were employed in the Goolsby laboratory to ensure that the samples provide reproducible results. The ability of the spectral instrument to improve spectral resolution will be tested by comparing the separation of multiple cell types in mixed samples using the spectral deconvolution and standard compensation matrix analysis methods. We will construct test samples composed of known mixtures of different cell types in different ratios and use these compare the standard and spectral deconvolution methods. Improved Separation of Blood Cells Project 1 is largely based on the ability of acoustic focusing techniques to manipulate particles in a flowing sample stream. We have already demonstrated that different sized particles in a mixture can be separated in the flow stream by adjusting the acoustic frequency and amplitude in the line drive (see Preliminary Data section in Project 1). We have also already demonstrated that the acoustic energy needed to separate mammalian cells is well below the threshold for causing physical damage to cellular membranes, meaning that we should be able to manipulate blood cells while maintaining their viability. The acoustic focusing instrument developed in Project 1 will be used to determine if we can separate red blood cells from white cells in a mixed flowing stream, based on the large difference in cell size between these two particles. As described in Specific Aim 3 of Project 1 (see Figure 13), our first approach will be to concentrate the larger white cells to the center of the sample stream, where they will be removed using a central tube while the red cells flow around the outside. Initial studies will be performed using routine blood samples obtained locally as well as from blood banks. It is important to note that only a high degree of purification is required: a small contamination of the white cell sample with red cells will not be a problem for subsequent flow cytometric analysis. If we cannot obtain sufficient purity with a single pass, we can use multiple passes through the acoustic focusing chamber. Once we have verified the ability to separate red and white cells, we will obtain clinical normal and hematopoietic malignancy samples as noted above from Dr. Goolsby's laboratory to determine if we can perform the separation on these samples. The fraction, and quality, of leukemic cells obtained via the acoustic separating procedure will be compared to samples prepared using standard red cell lysis techniques in order to verify improved cell recovery. Once the procedure has been demonstrated and optimized, it will be possible to implement a simple acoustic chamber for blood cell separation that could be used as a stand-alone preparative instrument or retrofit onto the front end of a standard flow analyzer.